Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 65k
    5. 20-GEMMASCOPE-RES-65K
    6. 365
    Prev
    Next
    INDEX
    Explanations

    references to specific names or terms related to mathematical concepts or entities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    references to the "Cox" entity

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    Cox, Groupon, Hughes, HS, Mann, Saxons, Hilton

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 20-gemmascope-res-65k
    Configuration
    google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    BackStack
    -0.66
    ExecuteReader
    -0.64
    شهاد
    -0.63
     ouvertes
    -0.57
     Arce
    -0.56
    iteness
    -0.55
    erir
    -0.54
    WriteLiteral
    -0.54
    OneToOne
    -0.53
     croire
    -0.52
    POSITIVE LOGITS
     متعلقه
    0.87
    ItemBackground
    0.81
     '{@
    0.81
    >{@
    0.74
    OGND
    0.73
     ujednoznacz
    0.72
     Grou
    0.72
    ruptedException
    0.71
    afficheront
    0.71
    styleType
    0.70
    Activations Density 0.091%

    No Known Activations